At the end of the last ice age, when 
the high glacial cliffs 
began to shrink back across a scarified continent, woodlands more typical of northern latitudes covered parts of what we now call Texas. Grassland savannah flourished, with pine and possibly aspen growing along streams and rivers. On the Llano Estacado, where today giant windmills sprout from endless cotton fields, there might have been substantial forests, or short-grass prairies, or a desert, or perhaps an open forest steppe—a grassy parkland with clumps of deciduous trees. Large Pleistocene mammals, such as mastodons, mammoths, camels, horses, and giant bison, grazed and wandered through seas of grass, hunted by dire wolves and saber-toothed cats. The coast, hundreds of feet lower and miles beyond the present shoreline, was drier, perhaps with wide dune structures, sand laid down by cold glacial winds blowing over a bare midwestern tundra.

The Trans-Pecos, now part of the vast Chihuahuan Desert, was temperate, with tall grasses and extensive open woodlands composed of piñon, juniper, and oak. As the glaciers receded and the oceans rose, the climate remained cool for a long time but grew progressively wetter, until it flipped and entered a long period of heat and drought, a trend that has continued, with some brief moist intervals and dramatically colder episodes, for the past nine thousand years.

Life in West Texas has never been easy, and to outsiders, the idea that people might choose to live in such country has always seemed improbable. In the sixteenth century, when the Spanish first passed through the area defined by the confluence of the Devils River, the Pecos, and the Rio Grande, they found little more than abandoned rancherías. When the Americans came through three centuries later, they saw these canyonlands as just another obstacle on the way to California—a desolate march between Fort Clark and Fort Lancaster best left to outsized characters like Jack Hays and Bigfoot Wallace. This was Indian country, and military maps all noted the presence of painted caves.

Cattlemen saw it differently. When the early Texas ranchers, among them my great-great-great-grandfather Perry Wilson, drove their livestock into the open rangeland along the Devils River, they found what appeared to be a stockman’s paradise. Cliffs along the Rio Grande and the Pecos made access to water difficult for livestock, but the Devils was easily approached and the grass was high and plentiful. What they didn’t know was that this paradise was dangerously fragile. Before long the tall grass began to fail as drifting cattle damaged the thin mantle of soil; immense herds of sheep soon followed, grazing down the short grasses, and when thunderstorms came the soil washed away.

Perry and his cattle stayed on the Devils for three years and then moved on; he was always a restless man, traveling back and forth to California, once by way of Panama, and he spent much of his life wandering along the western margins of Texas. In 1893 his oldest son, T.A., returned to the Devils River, near Juno, along with his wife, Bettie, and their young children. We still operate the ranch my ancestors built, but the rural society and economy that nurtured generations of my family is mostly gone, swept away by abstract economic forces and the ravages of a desert climate. By the time I was wandering on horseback through that country, the glory days of the sheep and goat industry were well behind us. Grasslands had given way to invasive mesquite and cedar, and drought was just a way of life.

The passage of the ranching world was swift. Not so the world of those who came before us. Enigmatic evidence of a human presence in the area reaches back millennia. At sites like Bonfire Shelter, the oldest and southernmost example in North America of a bison jump, and at Cueva Quebrada, another shelter, the butchered bones of Pleistocene mammals were found in deposits dated to some 14,000 years before the present. As rainfall gradually diminished over the centuries and the big game disappeared, people either adapted or kept moving. Those who stayed developed an ingenious hunter-gatherer economy based on desert plants and small animals and the occasional deer. Some left paintings as a record of their lives in this place.

Growing up here, I had little awareness of those ancient people, though signs of their presence were all around me. I played cowboys and Indians around their flintknapping sites and earth ovens and wickiup rings. I hunted for arrowheads. I had heard about Indian paintings, even saw some once or twice, but as a child I never paid them much attention. I had no idea that a complex of rock shelters just miles from my family’s ranch contained one of the most significant bodies of rock art in existence. Still less could I have suspected that those paintings, created for mysterious reasons some four thousand years ago, might speak to us today.

LISTEN to a short radio segment produced with KUT News and StateImpact Texas featuring Roger D. Hodge: 


I arrived at Seminole Canyon on a cool, windy morning in late March. The plan was to meet up with a group from Shumla, a center for archaeological research near Comstock, and spend a week studying the rock art of the Lower Pecos. I hoped to learn more about the deep history of the landscape in which I was raised, about the ways humans had tried, successfully and not, to live in it. Rain had been falling all across the state, and there was some hopeful speculation that the worst single-year drought in Texas history might be drawing to an end. I was doubtful; the previous year had been so dry that much of the cedar had died along with the grass.

I asked Elton Prewitt, a Shumla archaeologist, if he thought the drought was over. Paleoclimatology, it seemed, had some hard lessons to teach us. “What people don’t understand,” Elton said, “is that from around 8,500 to about 4,500 years ago, we had two back-to-back 2,000-year droughts out here that were much more severe than anything we have experienced in historic times.” These droughts were separated by a moist interval of perhaps a hundred years. “That’s the way it goes out here. You get a brief cool, moist period and you get hot and dry, with flashy spring-summer rains, which of course creates erosion and floods.” We talked about the big 1954 flood, a “once-in-over-10,000-years event in this country,” which scoured many of the local canyons but reached few of the major pictograph sites. The people of the Pecos River were no strangers to heavy weather.

Once our group of professional archaeologists and amateur rock art enthusiasts had gathered, we hiked down into Seminole Canyon, which feeds into the Rio Grande gorge a few miles to the south. Scarlet ocotillo blossoms stood out in pleasant contrast against gray skies and gray limestone. Clumps of green, thorny shrubs—blackbrush and catclaw, prickly pear and sotol—dotted rocky hillsides almost devoid of grass or anything resembling soil. Looking up and down the steep canyon, I saw two great cavities in massive limestone cliffs where water, cutting its meandering way through the sedimentary bedrock, had carved out spaces in which humans and other animals found shelter from the sun and wind. Early settlers and ranchers used such shelters as well, both as dwellings and as ready-made barns for livestock.

The first pictographs came into view as we approached Fate Bell Shelter, reached by a narrow trail along a jumble of boulders and small trees, Gregg ash and persimmon among them. Spilling out of the site was a talus slope of burned rock, untold centuries of household garbage. The largest paintings were huge—one was about 28 feet tall. When they were in full, vibrant color, about four thousand years ago, they would have been visible from across the canyon. Most of them had faded over the centuries, and dust from excavations and looting still clung to their surface, but many remained vivid and distinct. These were mostly done in what’s known as the Pecos River style, which features a dizzying variety of shapes and sizes. Figures of deer, a mountain lion with red lines radiating from its mouth, a winged anthropomorph sprouting antlers from his head, and strange ghostlike creatures covered the wall in what at first appeared to be a chaotic muddle, with images running into and over one another. Carolyn Boyd, Shumla’s executive director, pointed out the careful lines and the challenges that faced the ancient artists. Slowly, I began to see the planning and skill that must have gone into these paintings.

Interpreting ancient rock art requires concrete physical evidence as well as an openness to the possibilities of visual communication. The paint itself provides some clues. Using pigments derived from local minerals such as manganese, limonite, hematite, and ocher, the artists needed something to act as a binder, like egg white or blood. The most likely candidate, Carolyn believes, was animal fat—specifically deer marrow, mixed with a soapy liquid extracted from yucca root. That combination not only works, she said, but makes a gorgeous paint.

From the archaeology of the dry rock shelters, we also know quite a bit about what these people ate. By the time the Pecos River style appeared, the climate was far too dry for the buffalo. The people mostly lived off wild food plants: the sotol and lechuguilla they cooked in earth ovens, native onions and oregano, fruit like persimmons and grapes. They ground acorns and mesquite pods in their rock mortars, several of which we could see right here in Fate Bell. Their desiccated feces, known as coprolites, reveal that grasshoppers were a staple, as were minnows, snakes, lizards, and any other small creatures they could catch. They organized group hunts in which rabbits and deer were driven into snares or natural traps, but meat was not a large part of their diet. It was hard country, especially during the long dry spells.

“When I first started working out here,” Carolyn said, “I was told that the paintings were graffiti, or that it was something that they did only in their leisure time. Well, if they were using animal fat, they were sacrificing food off their plates, because fat was a precious commodity. In the desert you’re not going to find a lot of sources of fat, so that was a group sacrifice. When a deer was caught, which was rare, it was a communal sacrifice to make the paint. That shifts our understanding away from what you do in your leisure time to what you do before you do anything else.” It’s possible they spent all year collecting materials, making paint cakes to be used for some special festival or ceremony when the whole community would come together.

But who were these people? And why did they consider these paintings to be so important? For many years archaeologists had little to offer in the way of an explanation. The pictures were beautiful and intriguing, but they were difficult to understand, and no scientific methodology had been established for working out their meanings. One prominent theory was that the Pecos River–style pictographs were records of shamanic visions, hallucinogenic trances, or journeys to the spirit realm. Other researchers thought they depicted warfare. Many assumed the true meaning of the artwork was forever lost, that it had died with the culture that produced the paintings.

When Carolyn first encountered these images, in 1989, she was a professional artist fascinated with Indian paintings. She was unable to reconcile what she was reading about the pictographs with what she was seeing in the shelters. As an artist, it was obvious to her that the panels were conceived and executed as compositions; they weren’t just random indigenous graffiti or an anthology of individual shamanic visions. But she was equally sure that the archaeology profession would never listen to an outsider like her. So she went and got her doctorate in anthropology from Texas A&M and wrote her dissertation on the rock art of the Lower Pecos. Since then she and her colleagues at Shumla, which she founded in 1998, have transformed the study of these astonishing paintings.

Carolyn’s methods are painstaking. She has recorded the pictographs at dozens of sites, drawing them herself and repainting them to scale, quantifying their motifs, then comparing her data and the iconographic patterns that she has identified with the ethnographic record. What she and her team have discovered is that the Pecos River pictographs seem to be depicting certain archetypal myths of the great Mesoamerican civilizations: the Aztecs and the Mayans, and especially the Huichol, a culture whose religion has remained remarkably free of Christian influence to this day. The implications of this are dramatic: if Carolyn was right, the paintings were narratives and could be understood as North America’s oldest surviving texts.

Linguists and anthropologists have long surmised that some common linguistic and cultural ancestor must have existed to account for the profound similarities of language and religion among the Mesoamerican civilizations. They call the original cultural strain Proto-Uto-Aztecan. (Uto-Aztecan languages include those of the Aztecs and Huichol, as well as those of northern cultures such as the Ute, Hopi, Shoshone, and Comanche.) Many scholars believe that the Proto-Uto-Aztecan culture would have originated in what is now the southwestern United States, possibly in the Great Basin region (present-day Utah and Nevada), before spreading outward across the continent. According to Carolyn’s research, one strand of that migration appears to have passed through the Lower Pecos, leaving behind paintings that represent the archaic core of what eventually developed into the Mesoamerican religions. It may very well be the case, Carolyn told me, that “the origins of those belief systems were first documented in the rock art here on the Lower Pecos, four thousand years ago.”

Carolyn received powerful confirmation of her thesis in 1999, when she took one of her renderings of the so-called White Shaman panel to Real de Catorce, a town in northern Mexico. She showed the painting to a Huichol man there. “He became very quiet, and he said, ‘That’s our pilgrimage,’ and proceeded to tell the story of the peyote hunt using that panel as the storytelling device,” she said. Years later, in 2010, a Huichol shaman came to Shumla and told Carolyn that his people had traveled along the body of the great serpent, from its tail in the Great Basin to its head in the Valley of Mexico, and that the serpent passed through this place.

The central ritual of the Huichol religion is the peyote hunt: a pilgrimage to Wirikuta, the Huichol spiritual homeland in the Catorce Mountains, that reenacts the birth of the sun and the creation of the peyote cactus. The Huichol say that their ancestors, wolf people who were neither human nor fully animal, were led out of the cold, dark west and through the underworld by a deer, who gave himself up to sacrifice when they reached their destination in the east, Dawn Mountain. After the wolf people slew the deer, his body became peyote, which they ate, and the sun rose for the first time. According to Huichol traditions, peyote still grows in the tracks of the sacred deer, and they believe the life-giving rains follow the deer as well. In fact, after a rain, peyote buttons bulge with moisture and rise from the surface of the soil, often in the tiny cavities created by the hooves of grazing deer.

Indeed, for the Huichol, deer and peyote and water are all symbolically indistinguishable. Other stories tell of a Sacred Deer Person who flies down from the sky with peyote in his antlers. After setting in the west, the sun rides through the underworld on the back of a deer, gathering water to give to the people and the land when he reaches his zenith, during the summer rainy season. When the Huichol make their pilgrimage today, they do so to ensure that the sun will rise anew each morning and that he will continue to gather up rainwater during the night, rainwater without which the world would surely die.

Later that first day, after stowing our gear at the Shumla campus, we hiked past the remains of prehistoric earth ovens on the uplands to a small shelter known as Javelina Grid. On the way, Elton showed us how to recognize the signs of middens resulting from those ovens, mounds of earth and fractured, heat-darkened limestone. Cooking sotol and lechuguilla for several days to make the plants digestible required a huge investment of time and labor in return for a meager amount of nourishment. And yet for thousands of years these foods were a staple.

We picked our way along the slope of the canyon. Wild oregano grew thick and unbelievably fragrant alongside native tobacco. When we reached the shelter, we found room for just two or three people to crouch below the smoke-blackened ceiling. Javelina dung carpeted the ground inside. Several moments passed before I was able to grasp what I was seeing: an abstract grid pattern had been incised everywhere, all across the ceiling and walls of the shallow depression. Elton pointed out that similar sites are found up and down the Pecos. No one can say for sure who made these carvings or when, but the same patterns were found on incised stones at the Gault site in Central Texas, an important source of artifacts from the Clovis culture, once believed to have been the first to inhabit North America.

One of our group, a rock art specialist from Oregon named Jim Keyser, suggested that the art might represent “entoptic phenomena,” visual patterns that reflect the structure of the optic nerve. They show up in the early stages of a hallucinogenic trance. Unlike the massive figures we saw at Fate Bell, which were very public, communal, and probably required the construction of scaffolding, this art was private, best experienced by one or two individuals, perhaps as part of a vision quest.

Over the next week, as we hiked into shelters on the Devils River, on the Pecos, and in neighboring canyons, we saw much of this odd scratching, both on its own and among the more dramatic large-scale murals of the Pecos River style. At some sites the chipping, scratching, incising, or rubbing was everywhere; at others it was highly selective, as if the figure that was singled out was a special source of power or an object of fear. Often it was the antlered figure of the Sacred Deer Person, his tines adorned with peyote buttons, who was scratched most aggressively. Perhaps the paint chips were used in rituals to summon the rain, or to start new batches of paint for rituals to ensure the sunrise.

One day, after a harrowing night of thunder and wind threatened to carry off our tents, we all climbed into pickups and headed north of Comstock on Texas Highway 163. Leaving the blacktop, we drove through one ranch after another, carefully unlocking and locking gates as we bounced down fourteen miles of dirt roads. At last we came to the magnificent canyon of the lower Devils River, just above its confluence with Dead Mans Creek. We waded the river, stepping over deeper flutes cut in the smooth bedrock as monstrous carp coursed up and down the swift stream at our feet. Our destination lay miles upriver, so we walked carefully along the eroded limestone, among shapely grooves and terraces and overhangs; we examined veins of chert, and Jim explained the techniques used by the Indians to mine the glassy silicate mineral used for knives, scrapers, and projectile points.

We turned from the riverbed and climbed a jagged slope of boulders and faulted caprock to the upland, where we encountered a ranch road and signs of a more contemporary economy: a small shed, several deer blinds, and dozens of spent shells. Resurrection plants, dormant until the recent rains, were sprouting up along our trail; prickly pear and strawberry cactus blooms vied for my attention with yellow, red, and blue wildflowers. We carefully descended a narrow trail along the rim of Cedar Springs Canyon and entered the shelter in silence as the roar of a passing military jet drowned out the melodic calls of wrens and sparrows in the thick brush nearby.

As the sound of the jet faded into the distance, I began to discern the patterns on the wall. Small, delicate depictions of felines and humans and deer—often with exquisitely drawn antlers; sometimes with puckered lips—were intertwined with a thick fabric of somewhat larger figures: men carrying atlatls and spears, a mountain lion spitting blood or speech or medicine, ghostly outlines of humans or gods, ambiguous designs resembling something from Klee or Kandinsky or Miró. Almost all the larger images bristled with hairlike protrusions; many were impaled with spears. Carolyn told us that the paintings here had been dated to 3,800 years before the present, and she zeroed in on one particular image, which shows up repeatedly across the region. Tall, heavily incised, and painted in red, yellow, black, and white, it was a vaguely triangular figure with arms protruding from the upper corners and a small head with an open mouth resting on the line between them. A pair of deer antlers emerged from the head, and two long, serpentine lines projected from the shoulders, encompassing a mysterious black square and linking up to a yellow circle. A field of radiating lines filled the space between the circle and the square. Almost everything in the panel, in one way or another, seemed to be connected to this figure. This was, Carolyn said, “the point of intersection between this world and the other world”—the axis mundi, the mouth of the gateway serpent.

Years before, Carolyn was camping out here with students from A&M, recording this site, when a fast-moving thunderstorm blew in. They had been swimming down by the river and had to run for cover, just making it to the shelter when the skies opened and sheets of water began pouring over the edge of the cliff. “It was like being behind a waterfall; the thunder was rolling, and the bedrock was just shaking,” she recalled. “You could feel it through your entire body.” Sitting there, behind that sudden wall of water, protected from the rain and the wind and the lightning in a place that had been sacred to the people of this desert for thousands of years, Carolyn and her students eventually drifted off to sleep. When they awoke, the ancient basin below the paintings was full, holding water just as it had done thousands of years before. Desert sunlight reflected off the water onto the walls, and the paintings came alive with a motion of limbs.

No one knows exactly what happened to the great artists of the ancient world who left us these pictographs. Their tradition persisted for more than a thousand years, and then one day they put down their brushes, gathered their children, and left. Others came and lived in much the same way, and some left their images on the walls as well. When the soldiers and railroad workers arrived, they showed little interest in the paintings, often simply adding their own names to the walls or using them for target practice.

As we hiked down into White Shaman shelter, a site a few miles above the mouth of the Pecos, I thought about the continuous migration of these various peoples, some known, some unknown. The same forces had driven one group after another into and then out of these lands. I wondered if future archaeologists would puzzle over our rusty implements and spent shell casings.

White Shaman lies in a steep gorge that feeds into the main channel of the river among imposing limestone cliffs. We climbed up and down a steep trail and eventually arrived at the shelter, surrounded by a very high fence topped with three strands of barbed wire to discourage vandals. Days of clambering up and down such cliffs had begun to take a toll on my knees, but my exhaustion fell away when I saw the paintings.

All the main characters from the Huichol creation stories were there: impaled deer covered with black peyote dots; five black ancestor figures spread across the panel, carrying ten candles as offerings; a headless white moon goddess (the White Shaman), sometimes known as Mother West Water, facing due west; an antlered figure emerging from Dawn Mountain; and Grandmother Growth, an earth monster who appears as a combination of serpent and catfish. There was even a figure of the first man, who was saved by the moon goddess from a great flood, a tiny human figure suspended above a canoe. Peyote buttons were ritually impaled in the pictographs just as the Huichol continue to do on their pilgrimage. It wasn’t at all hard to imagine that the artists who painted this panel, like today’s Huichol pilgrims, believed that their own sacrifices—of precious deer marrow as well as energy and time that might otherwise have been invested in gathering food—were necessary to guide the sun through the underworld each night, riding on the back of a deer to be reborn in the east, so that the rains might follow and bring forth the desert’s fruits, among them the sacred peyote.

But perhaps one spring the painting failed. Perhaps the rains did not come for several years or more. And perhaps the people came to see a long drought as a sign to move on to the south—where Grandmother Growth was more generous, where the Sacred Deer Person was more reliable. We don’t know exactly what happened. “What we’re learning from studying the rock art,” Carolyn explained, “is that they followed the rain—they followed the water.” Nothing was more important to these people than water. In that sense, the experience of the Pecos River people was probably not so different from our own. Like us, they faced violent weather and climate change and drought. Like us, they believed the existence of the world depended on their actions. “Everything follows the rain,” Carolyn said. “Everything follows the water.”

Thousands of years after the paint dried on the rock art of the Pecos River people, restless souls like my Wilson ancestors grazed their livestock along the Devils River. Some of them settled and raised their families, and soon a town appeared, with saloons and hotels, a post office and a school. Time passed and most people moved away. Little remains of that brief episode but a handful of scattered ranches that haven’t been sold to oil tycoons or tobacco lawyers. Compared with the long tenure of the Indians, the European presence here has been as fleeting as a cloud of dust.

Still, some things don’t change. At Painted Shelter, on a ranch owned by Missy and Jack Harrington, who donated the land for the Shumla campus, a stream flows along the base of walls that bear the faint remnants of Pecos River–style pictographs. The images, ancient and faded and almost illegible, lie beneath large paintings that are much younger but still pretty old, examples of what is called the red monochrome style. Just up the hill stand the ruins of a small house constructed from native flagstones that was already present when Missy’s great-grandfather settled here. Inside that old structure, near the floor, a man named Frank Greenwood carved his name and his family’s brand in 1891, possibly while working as a U.S. marshal. Greenwood’s granddaughter Frankie Lee was one of my grandmother’s oldest friends. Missy used to picnic along the bank and swim in that stream when she was a child. For generations, people have been drawn to the cool shadows of this place. They came, rested, and passed on, leaving their marks, their signs and symbols, on the shelter walls.